Just sneaking in by the side door here (as aircrew); one of the frustrating aspects of the techie/aircrew interaction was the latter knowing that a fault had been experienced and the former carrying out the fault finding process and not finding said malfunction - thus "NFF" in the 700.
Subsequent sorties and the fault continues to manifest itself and remains undectable on the ground; hence a dgree of frustration all round.
One that particularly springs to mind was an avionics/electrical fault on the Vulcan, which was eventually resolved by the affected nav sitting down with the split brain Ch Tech and describing in detail what happened and the stage of flight at which the fault materialised. Seems the cabin pressuristaion deformed the rear bulkhead by a tiny amount, and the cables connecting cabin eqpt to electrical supply were ever so much too short, causing them to ease out of the connector by an amount sufficient to break the connection, then subsequently get pushed back in when the pressurisation process was reversed.
Similar symptoms on the next generation - Tornado - where airborne malfunctions could also not be reproduced on the ground; only to be expected when the dynamics of flight are imposed on the the sensitive parts of the kit.
So, not trying to do the engineer's job, but reporting as accurately as possible the fault and the conditions of flight when it appeared. Not even suggesting any suspicions as to cause (it would only be guesswork, given the aircrew limited knowledge of the systems they operate regularly...), but giving as much detail as might be required for techie head scratching to achieve a solution - better than just changing the LRU and experiencing the fault on subsequent flights. I believe we used to call it teamwork (except that didn't work too well when we had centralised servicing - much better to all be of one squadron).
Mister B